Well-kept secrets of Zope

Martijn Faassen

2007-09-10 15:24

Well-kept secrets of Zope

Zope is a web framework that comes equipped with powerful, apparently secret, features. Some of the things Zope has had for literally years other web frameworks are only evolving today. And in other cases, Zope comes equipped with features that other web framework communities are currently only dreaming about.

If Zope has such powerful features, why are they a secret? I think it's a combination of an unfortunate (but partially well-deserved) reputation Zope has in large parts of the Python community, and Zope isolating itself in its own community (it managed to build a large community many years ago).

Before I start off yet another discussion with people burned by Zope 2 in the past: Zope 3 is not Zope 2. It's not crufty. It is hard to approach, but we're fixing this with Grok, which cuts down on the complexity hard.

Now let's go into some features that Zope has and that we seem to be keeping a secret very carefully from the rest of the world.

Zope 3 has unicode everywhere

In july 2007, the Django project started to support unicode. I'm happy to say that the Zope 3 project has supported unicode throughout since 2002. Consistently. Without backwards compatibility cruft. When the Django project gains this support, I, who doesn't really follow Django developments, learned about it from various blogs. We in the Zope community don't tell anyone as that'd spoil the surprise. Or something.

Zope 3 has a built-in form generation and validation system

Last year, I heard a lot about ToscaWidgets. This year I hear a lot about Django newforms, replacing Django's older form generation system.

Development of Zope 3's declarative schema system and form generation system started, guess when, in 2002, based on ideas we had been exploring for a few years by then in the Zope 2 context. We've had 5 years of experience with it since then. This led to the evolution of a new form generation system (on top of the existing, solid, declarative schema system) in 2004. This year we've seen a further evolution of this system with z3c.form (a fourth new forms iteration evolving the work that had gone before).

So, Zope has a headstart of years of experience. We keep this hidden within our own community, because otherwise it'd be like, telling!

An object database

CouchDb is gaining some attention recently. While not an object database, it promises to store documents, not relational database tables. Recently we've also witnessed some ORM Wars in various Python blogs.

Meanwhile, some people in the Rails world have been thinking... Wouldn't it be nice if instead of all this impedance mismatch between objects and relational databases we could use a true object database? Wouldn't that be cool?

It's time for me to yawn and say "been there, done that". People might somehow have missed it, but Zope 3 is equipped with an ACID-compliant, clusterable object database, the ZODB, that has been under development since 1997. The Zope people know the benefits of document-oriented, object databases for web applications. We've worked in this impedance-mismatch-free world for so long that we know the drawbacks too, and thus have built Zope 3 extensions to work with SQLAlchemy as well.

Buildout

We have been putting together complicated web applications from many different bits and pieces for a while now in the Zope world. Setting up a development environment or rolling out a deployment is quite an involved job often involving endless INSTALL.txt files. Setuptools and eggs offer a lot of help here, and the Zope project has been embracing them in a major way. They still leave a lot of stuff to do by hand however, especially if non-library components are involved such as web servers.

Enter zc.buildout. zc.buildout is an extensible Python system for assembling Python applications from multiple parts. It will work for any Python project: I've used it with TurboGears, and to set up a game development environment, among others. Zope itself is of course a major user of it, and it was forged in the fires of long experience and requirements of the Zope community. It's a lot to chew on to learn, but I believe any significant Python development project can benefit from using it. I believe buildout is another example of where the Zope project is tackling problems years ahead of many others within the Python community. And now we've told you about it.

To conclude

These features shouldn't be a secret. We should shout it off the rooftops: the Zope community in many ways is still pushing the frontiers of Python web framework development.

Zope 3 has powerful features, and now also has an easy entry point. If you want an easy way to learn about the future of web development in the Python world, please try it out in the form of Grok, which promises to make Zope 3 safe for the cave man or woman in all of us.